Flu shot guidelines criticized
Mathematical model suggests that US experts got their priorities wrong.
The US policy for which groups should be the first to receive influenza vaccines is not the most effective strategy to limit the spread of swine flu, according to a paper published online today in Science1.
Last month, a US Centers for Disease Control and Prevention (CDC) advisory panel recommended that young people aged from 6 months to 24 years old should be placed at the front of the queue for flu jabs. But a new mathematical model shows that very young children do not contribute greatly to transmission, and priority should be given to school-aged kids, the number one cohort of disease spreaders, as well as their parents, who serve as gateways to the rest of the population.
"Those vaccines could be better used in older kids and the parents of older kids," says Jan Medlock, a mathematical biologist at Clemson University in South Carolina, and an author of the study. "Contacts go way up once kids reach five and go to school and are in the big pool with the big kids."
CDC spokesman Tom Skinner says that the agency will closely consider the model's findings, but doesn't plan to adjust its guidelines. "Our immunization recommendations really are not based on one single model," he says. "Models are based on many assumptions. They can be a helpful tool for identifying possibilities, but it's really not possible to know how any specific model might play out."
Medlock, together with Alison Galvani, an epidemiologist at Yale University in New Haven, Connecticut, developed an age-structured model of influenza dynamics based on mortality data from the 1918 and 1957 flu pandemics, infectivity data from the latest pandemic H1N1 outbreak, and recent survey data of how often people contact each other. The best way to dole out flu shots, they found, was almost always to prioritize people aged 5-19 years and 30-39 years. This was true when success was measured in different ways, including limiting deaths or infections, and minimizing years of lost life or economic costs.
"If you vaccinate the kids and you vaccinate the parents then you can virtually terminate transmission to the rest of the population," says Galvani, who is funded by the National Institute of Health's Models of Infectious Disease Agent Study (MIDAS) programme, which supports computational and mathematical research to control infectious diseases. On 18 August, Galvani met with Robin Robinson, director of the Biomedical Advanced Research Development Authority (BARDA) in Washington DC, to discuss the paper's conclusions as part of an internal working group-meeting between BARDA and MIDAS.
William Schaffner, an influenza expert at Vanderbilt University Medical Center in Nashville, Tennessee, says that the paper confirms and validates the CDC's focus on vaccinating schoolchildren. "As far as I'm concerned, the glass is way more than half full with this," he says. Indeed, Medlock says that the CDC is on the right track by moving away from prioritizing elderly people, as it has in past seasonal flu policies. "Our only real quibble would be using vaccines on children under five," he says.
But the model doesn't agree with the CDC's guidelines for parents either. The latest recommendations call for swine flu vaccines for parents with children younger than six months old to create a "cocoon" of protection around vulnerable babies. But Medlock and Galvani's model indicates that vaccinating parents of children aged five or older would block transmission more effectively. "That's something that's not necessarily intuitive," says Kumanan Wilson, who studies public-health policy at the University of Ottawa in Canada.
Wilson says that diverting the focus away from kids under five might also be "more feasible" than the CDC's 6-months-and-up policy because parents are often wary of vaccinating young children.
The CDC's swine flu vaccine priority list also includes pregnant women, health-care workers, and non-elderly adults with high-risk medical conditions. None of these groups were included in the strictly age-based model.
The World Health Organization recommends a step-wise approach to prioritizing swine flu vaccine handouts for healthy individuals: 15–49 year-olds top the list, followed by children, older adults and elderly people. Medlock says that he didn't explicitly model these guidelines, but that unless you have very little vaccine stock before the epidemic begins, it's probably not the best strategy.
Many national governments have not issued policies for vaccinating healthy people. For example, the UK Department of Health says that priority should be given to at-risk groups, and has not suggested focusing on any particular age group for healthy people.
Medlock and Galvani's model "is best taken as one piece of work among many", says Irene Eckstrand, scientific director of MIDAS. "I'm cautious about how it applies to policy." Models can inform decisions, but they don't take into account many real-world goals beyond simply halting transmission, she says. That's why public-health officials have to weigh up all the priorities. "Otherwise, we'd have a room full of mathematicians with a computer determining public policy," says Schaffner.
- Medlock, J. & Galvani, A. P. Science advance online publication doi: 10.1126/science.1175570 (2009).